96802 - Data Mining for Business and Market Research

Learning outcomes

This course will present the main data mining methods used in knowledge discovery in business employing internal and external data. With an emphasis on data analysis and on the use of a software special attention will be devoted to techniques that help to single out the relationships of interdependence and patterns in business and market research phenomena. Students will learn, hands-on, how to organize and analyse market research data. In particular, at the end of the course students will be able to: - independently run a complete data mining process (from data pre-processing to the interpretation of obtained results); - choose the best suited statistical methodology for the problem at hand; - to critically interpret empirical results.

Course contents

This course provides an advanced introduction to the statistical and data mining techniques used for knowledge discovery from business and market research data. It combines methodological foundations with extensive hands-on applications, enabling students to develop the analytical skills required to transform complex datasets into actionable business knowledge.

Students are introduced to the complete data mining process, from understanding the business problem and preparing the data to selecting appropriate analytical methods, interpreting statistical outputs and communicating evidence-based conclusions. Throughout the course, particular emphasis is placed on methodological reasoning, critical interpretation of results and the ability to justify analytical choices in realistic business contexts.

The course combines theoretical lectures with extensive laboratory sessions based on SAS, where students analyse real business and market research datasets. Practical activities progressively develop students' ability to perform an entire data mining project independently, strengthening both technical competence and statistical reasoning.

Although previous knowledge of introductory statistics is assumed, no previous experience with SAS or data mining software is required.

The course is organised into two learning units, each combining theoretical lectures and laboratory activities.

---------------------------------------

Learning Unit 1 (Weeks 1–5)

 Introduction to Data Mining

• Data-analytic thinking.
• The Knowledge Discovery in Databases (KDD) process.
• From business problems to data mining tasks.
• Real-world business applications.

Data Preparation

• Data objects and attribute types.
• Data matrices and transformations.
• Data cleaning.
• Missing data.
• Data preprocessing.

Statistical Computing with SAS

• Introduction to SAS.
• Data organisation.
• Data management.
• Data preprocessing using real datasets.

Dimensionality Reduction

• Principal Component Analysis.
• PCA for ranked variables.
• Multiple Correspondence Analysis.
• Interpretation of latent dimensions.

Learning activities

Students progressively develop a complete data preparation workflow using SAS, explore real business datasets, apply dimensionality reduction techniques and interpret multivariate statistical outputs through guided laboratory sessions and classroom discussions.

Key milestone

At the end of Week 5, students may take the first mid-term examination, covering all topics included in Learning Unit 1.

-------------------------------------------

Learning Unit 2 (Weeks 6–10) 

Topics

Proximity Measures

• Similarity and distance measures for mixed data.

Clustering

• Hierarchical clustering.
• Partitioning methods.
• Hybrid clustering.
• Cluster validation.
• Interpretation of clustering solutions.

Profiling

• Behavioural segmentation.
• Customer profiling.

Association Rule Mining

• Support.
• Confidence.
• Lift.
• Interpretation of association rules.
• Business applications.

Predictive Analytics

• Data Mining Scoring.
• Model evaluation.

Causal Machine Learning

• Introduction to causal inference.
• Causal Machine Learning methods.
• Evaluation of marketing interventions.
• Interpretation of heterogeneous treatment effects.
• Decision-support applications.

Learning activities

Students analyse authentic business and market research datasets using SAS, compare alternative clustering solutions, discover behavioural patterns through association rule mining, build customer profiles and critically interpret predictive and causal machine learning results in realistic business scenarios.

Key milestones

• Completion of SAS laboratory activities throughout the course.
• Interpretation of statistical outputs and presentation of analytical findings during laboratory sessions.
• Discussion of representative examination questions during revision sessions.
• Students may complete the assessment either through the two mid-term examinations or by taking the final comprehensive examination, according to the assessment options described in the Assessment Methods section.

Readings/Bibliography

Required readings

The course is supported by both the required textbooks and the teaching materials made available on the University's Virtuale platform. These complementary resources are designed to be used together throughout the course.

Main textbook

Tufféry, S. (2011). Data Mining and Statistics for Decision Making. John Wiley & Sons.

Required chapters: 1–3, 7, 9–10 and 12.

This textbook provides the methodological foundations of the course and should be used to consolidate and deepen the concepts introduced during lectures and laboratory sessions.

Causal Machine Learning

Hernán, M. A., & Robins, J. M. (2020). Causal Inference: What If.

Available free of charge at:

https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/

Selected chapters will be assigned during the second part of the course when causal inference and Causal Machine Learning methods are introduced.

Recommended readings

The following papers provide additional methodological insights into causal inference and matching methods and are recommended for students wishing to deepen their understanding of modern evaluation techniques.

• Becker, S. O., & Ichino, A. (2002). Estimation of average treatment effects based on propensity scores. The Stata Journal, 2(4), 358–377.
• Dehejia, R. H., & Wahba, S. (1999). Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs. Journal of the American Statistical Association, 94, 1053–1062.
• Dehejia, R. H., & Wahba, S. (2002). Propensity score matching methods for nonexperimental causal studies. Review of Economics and Statistics, 84(1), 151–161.

These readings are not required for the examination but provide useful background for students interested in advanced applications of Causal Machine Learning and programme evaluation.

Teaching materials

The teaching materials made available on Virtuale include:

• lecture slides;
• structured lecture notes;
• SAS laboratory notes;
• annotated SAS code;
• business and market research datasets;
• practical exercises;
• sample examination questions;
• additional reading material.

The lecture notes provide structured summaries of the textbook chapters covered during the course and are enriched with additional explanations, practical examples, SAS code and case studies discussed during the lectures. They are intended to complement, rather than replace, the required textbooks.

How to use the course resources

Students are encouraged to prepare for each learning unit by reading the relevant textbook chapters before or shortly after the corresponding lectures. The lecture notes and laboratory materials should then be used to consolidate the theoretical concepts through practical implementation in SAS and the interpretation of statistical outputs.

The recommended textbooks provide the theoretical foundations of the course, while the teaching materials available on Virtuale bridge theory and practice by illustrating how the methods are applied to real business and market research problems.

Students are encouraged to use both resources throughout the semester to support independent learning and preparation for both the mid-term and final examinations.

Teaching methods

The course combines interactive lectures with hands-on SAS laboratory sessions, providing students with the opportunity to progressively develop both methodological understanding and practical data mining skills.

The theoretical lectures introduce the statistical and machine learning methods underlying the data mining process. Rather than presenting algorithms in isolation, the course emphasises the rationale behind each analytical technique, its assumptions, strengths and limitations, and the types of business problems for which it is most appropriate.

Laboratory sessions are fully integrated with the lectures and are based on the analysis of authentic business and market research datasets using SAS. During these sessions, students progressively implement the complete data mining workflow, from data preparation to the interpretation and communication of analytical results. Practical activities are designed not only to develop technical proficiency with the software but also to strengthen students' ability to justify methodological choices and critically evaluate empirical findings.

Throughout the course, students are regularly asked to interpret statistical outputs, compare alternative analytical approaches and discuss the managerial implications of their results. Particular emphasis is placed on communicating statistical evidence using appropriate disciplinary language and on translating analytical findings into meaningful conclusions for business decision-making.

To support continuous learning, students are encouraged to review the recommended readings before each learning unit, actively participate in classroom discussions and laboratory activities, and complete the proposed practical exercises independently between classes. Home assignments are provided throughout the semester to consolidate the methods introduced during lectures and to allow students to monitor their own progress. Whenever appropriate, solutions and feedback are discussed during subsequent laboratory sessions.

Teaching materials, SAS code, datasets, laboratory notes, additional readings and sample examination questions are progressively made available through the University's Virtuale platform. Students are expected to consult the platform regularly, as it serves as the official repository for all course materials and announcements.

In view of the teaching methods adopted, attendance requires prior completion of Modules 1 and 2 of the University's online Health and Safety training, in accordance with the University regulations for computer laboratory activities.

[https://elearning-sicurezza.unibo.it/]

Assessment methods

The same assessment methods and evaluation criteria apply to both attending and non-attending students. Attendance is not compulsory; however, active participation in lectures and laboratory sessions is strongly recommended, as classroom discussions, SAS practical sessions and formative exercises are designed to progressively develop the analytical, methodological and interpretative skills assessed in the examination.

Students may complete the assessment either by taking two mid-term examinations or by sitting a single comprehensive final examination. The first mid-term examination covers the contents of Learning Unit 1 (Weeks 1–5), while the second mid-term examination, held during the first official examination session, covers the contents of Learning Unit 2 (Weeks 6–10). Students who choose not to take the mid-term examinations may instead take the comprehensive final examination covering the entire course syllabus.

The structure of the examination is identical for both the mid-term examinations and the comprehensive final examination.

Each examination consists of three parts:

• one open-ended theoretical question (40% of the final mark);
• one open-ended question requiring the interpretation and critical discussion of statistical outputs (20% of the final mark);
• one practical SAS exercise requiring students to independently perform a statistical analysis (40% of the final mark).

The examination is designed to assess whether students have achieved the intended learning outcomes of the course and are able to independently manage the main stages of a data mining project, from methodological reasoning to the interpretation of empirical evidence.

The theoretical question assesses students' understanding of the methodological foundations of the techniques introduced during the course. Students are expected to demonstrate knowledge of the statistical and machine learning methods, explain their assumptions, strengths and limitations, compare alternative analytical approaches, and justify the selection of the most appropriate methodology for different business and market research problems.

The statistical output interpretation question evaluates students' ability to critically interpret the results produced by statistical analyses, explain their substantive meaning, assess the quality of the analytical findings and translate quantitative evidence into meaningful business conclusions using appropriate statistical terminology.

The practical SAS exercise assesses students' ability to independently perform the required statistical analysis using SAS. Students are expected to correctly organise and analyse the data, implement the appropriate procedures, interpret the resulting outputs and justify the analytical choices made throughout the analysis.

Successful performance therefore requires both methodological understanding and practical competence. Students are expected not only to correctly apply the analytical methods introduced during the course, but also to demonstrate sound statistical reasoning, critically interpret empirical results and effectively communicate analytical conclusions in business and market research contexts.

Representative examination questions and practical exercises will be discussed during the course and made available through the University's Virtuale platform.

The examination is closed-book. Books, lecture notes, online resources, mobile phones, smart devices and any other unauthorised material may not be used during the examination.

Resit policy

Students who do not pass the examination or who wish to improve their mark may take the examination during any of the official examination sessions scheduled by the Degree Programme, in accordance with the University of Bologna regulations.

In accordance with the University Teaching Regulations, a passing grade may be refused only twice. If a student declines a passing grade, the result obtained in the subsequent examination attempt—whether higher or lower—will become the official recorded grade.

The same assessment format, learning outcomes and evaluation criteria apply to all examination sessions.

Evaluation criteria

The final mark reflects the student's level of methodological understanding, statistical reasoning, practical competence in SAS and ability to interpret and communicate quantitative evidence in business and market research contexts.

• Below 18 (Fail): the intended learning outcomes have not been achieved. The student demonstrates insufficient understanding of the methodological foundations of data mining and is unable to correctly apply or interpret the analytical techniques introduced during the course.
• 18–23 (Satisfactory): the student demonstrates a satisfactory understanding of the main concepts and methods, correctly performs standard analyses and provides basic interpretations of statistical outputs, although with limited critical discussion or methodological justification.
• 24–27 (Good): the student demonstrates good knowledge of the course contents, correctly applies the analytical methods using SAS, appropriately interprets statistical outputs and justifies methodological choices with only minor inaccuracies.
• 28–30 (Very Good): the student demonstrates thorough methodological understanding, confidently applies data mining techniques using SAS, critically interprets empirical results and provides well-reasoned business conclusions.
• 30 cum laude (Excellent): the student demonstrates outstanding mastery of the complete data mining process, combines methodological rigour with excellent practical competence in SAS, critically evaluates alternative analytical strategies and communicates statistical evidence clearly, accurately and convincingly.

Use of Artificial Intelligence

For the assessment, the use of generative Artificial Intelligence (AI) is not permitted. Any unauthorised use constitutes a violation of academic integrity and will be treated in accordance with the University of Bologna regulations

Students with disabilities or specific learning disorders (DSA)

Students with temporary or permanent disabilities or specific learning disorders (DSA) are encouraged to contact the University's dedicated support service well in advance. Appropriate accommodations may be arranged in accordance with University procedures and are subject to the approval of the course instructor, taking into account the intended learning outcomes of the course.

Teaching tools

The Virtuale platform serves as the official online learning environment for the course and should be consulted regularly throughout the semester. It is the primary repository for all teaching materials, laboratory resources, course announcements and examination information. The following resources will be progressively made available through Virtuale:

• lecture slides;
• structured lecture notes summarising the main theoretical concepts discussed during the lectures;
• SAS laboratory notes and annotated SAS code;
• business and market research datasets used during laboratory sessions;
• practical exercises and home assignments;
• sample examination questions;
• additional reading materials and methodological references.

Students are expected to use the recommended textbooks together with the teaching materials available on Virtuale. The lecture notes provide structured summaries of the textbook chapters covered during the course and are complemented by additional explanations, worked examples, SAS implementations and discussions of real business applications. These complementary resources are designed to support independent learning, laboratory activities and examination preparation.

The course makes extensive use of SAS for data management, statistical analysis and the implementation of data mining and machine learning methods. During laboratory sessions, students progressively develop practical skills in importing, preparing and analysing data, implementing appropriate analytical procedures, interpreting statistical outputs and communicating empirical findings. Students are therefore expected to become familiar with the SAS environment and to practise independently using the examples, datasets and laboratory materials provided throughout the course.

Virtuale also provides sample examination questions and additional practical exercises that enable students to monitor their progress throughout the semester and prepare effectively for both the mid-term and final examinations.

The practical use of SAS throughout the course is intended not only to support the understanding of statistical and data mining  methods, but also to develop the computational, analytical and problem-solving skills required to independently conduct data mining projects and address real-world business and market research problems.

Office hours

See the website of Ida D'Attoma